Quantitative measurement of aperture settings for generic camera calibration

Abstract

Abstract We investigated the effect of camera aperture settings on the accuracy of generic camera calibration, extending previous work with a larger measurement range and slightly modified methods. Four hardware configurations were built: (I) a monitor-based intrinsic calibration setup, (II) an integrating-sphere illumination system for quantitative aperture measurement, (III) an extrinsic-calibration jig with precisely milled reference plates and ArUco markers, and (IV) a validation setup containing two colored spheres of known spacing. We measured nine different aperture sizes and evaluated them for generic intrinsic calibration performance. We found no significant relationship between the aperture opening and the intrinsic calibration accuracy. After an ICP-based (iterative closest point) generic extrinsic calibration, we evaluated the distances between two colored spheres for validation and thus demonstrated the suitability of this method for such measurements.